Herpetological Conservation and Biology 6(3):364–371 Submitted: 5 April 2011; Accepted: 1 November 2011; Published: 31 December 2011. ECOLOGY OF A POPULATION OF THE EARTHSNAKE CONOPSIS BISERIALIS in the MEXICAN TRANSVOLCANIC AXIS 1 1,2 3 OLIVA CASTAÑEDA-GONZALEZ , JAVIER MANJARREZ , IRENE GOYENECHEA , AND VÍCTOR 4 FAJARDO 1Centro de Investigación en Recursos Bióticos, Universidad Autónoma del Estado de México, Toluca Estado de México, CP 50000, México 2Corresponding author, e-mail: [email protected] 3Centro de Investigaciones Biológicas, Universidad Autónoma del Estado de Hidalgo, Pachuca, Hidalgo, CP 42001, México 4Laboratorio de Ecología de la Conducta, Facultad de Ciencias, Universidad Autónoma del Estado de México, Toluca Estado de México, CP 50000, México Abstract.—Information about the ecology of Mexican snakes, such as Conopsis biserialis, is minimal. This species has low population densities and may be in decline over much of its historical range, apparently caused by destruction and fragmentation of the temperate forest habitat in central Mexico. We describe general life-history traits of this specialist fossorial species in a natural population from the State of Mexico. We captured 84 snakes from February 2003 to December 2004. Conopsis biserialis has a unimodal abundance pattern of eight to nine months with a peak in the rainy season (June to September). Of the 84 snakes we found, males and females have the same body size, mass, cloacal temperature, and a similar number of ventral scales, but males have longer tails, heads, and more caudal scales than females. We observed few neonates or juveniles during our surveys, and only one adult female was gravid with four palpable embryos. The sex ratio of adults was close to 1:1. We found 96% of snakes under rocks with a cloacal temperature 6° C greater than rock temperature. Our data suggest a specialized diet of ground-burrowing invertebrate prey (arthropod larvae). These data are broadly consistent with information from earlier studies on C. biserialis and other fossorial species, but are insufficient to detail the ecological characteristics of reproduction in this snake. The ability of C. biserialis to persist is constrained by anthropogenically altered habitat in central Mexico. We need to develop appropriate techniques for the study of these fossorial species to learn enough about these animals to plan effectively for their conservation. Key Words.—abundance; Conopsis; diet; ecology; microhabitat; reproduction; State of Mexico INTRODUCTION biserialis are isolated and not unified. Uribe-Peña et al. (1999) reported the use of the bunchgrass Mulenbergia Ecological information on snake species from the macroura as a possible microhabitat, whereas litter and Mexican Transvolcanic Axis is limited (Seigel and ground burrows were reported by Goyenechea (2000). Collins 1993). Studies on this topic for snakes in Conopsis biserialis eats beetles, moths, butterflies, and Mexico are rare, especially when compared to the occasionally spiders (Uribe-Peña et al. 1999). Data on numerous studies on snakes in the northern United States reproduction are also limited. It is known to be and Canada (Seigel and Collins 1993). Field data biased viviparous and mates in August, but further data are toward the northern species limit our knowledge about lacking (Greer 1966; Fitch 1970; Uribe-Peña et al. 1999; ecology of Mexican snake species. Most small fossorial Estrada-Virgen and Alvarado-Díaz 2003). This species snakes remain unstudied because they tend to be highly may be in decline over much of its historical range, secretive, cryptic, and infrequently active (Seigel 1993; apparently caused by destruction and fragmentation of How and Shine 1999; Goodyear and Pianka 2008). the temperate forest habitat in central Mexico (Flores- The genus Conopsis Günther, 1858, is a group of Villela and Gerez 1994). Small fossorial species may be specialist fossorial snakes (Greene 1997) composed of at particular risk from this anthropogenic disturbance. five morphologically variable Mexican species These kinds of snakes are poorly suited to moving long characterized by taxonomic instability (Goyenechea distances across open areas, especially across artificial 2000; Goyenechea and Flores-Villela 2002, 2006). The substrates that preclude burrowing (How and Shine Earthsnake, Conopsis biserialis (Fig. 1), is the most 1999). Thus, small fossorial species may be strongly taxonomically stable species compared with the other disadvantaged by habitat fragmentation. species of the genus (Goyenechea and Flores-Villela In our work, we focused on increasing the knowledge 2006), but the ecological characteristics of this species of an endemic and poorly known fossorial Mexican have not been well studied. The existing data for C. species located within an area highly disturbed by the Copyright © 2011. Oliva Castaneda-Gonzalez. All rights reserved. 364 Herpetological Conservation and Biology We weighed snakes with a precision balance (± 0.01 g) and measured snout-vent length (SVL) and tail length (TL) using a metric scale (mm). We measured head length and head width with digital calipers. We also counted the number of ventral and caudal scales on each snake (Rossman et al. 1996). We determined the sex of each adult snake by the thickness of the tail (Conant and Collins 1998) and by hemipenis eversion in newborn (< 96 mm SVL, sensu Goyenechea 2000) and juvenile (96–185 mm SVL) snakes. We counted the number of embryos by ventral palpation of pregnant females and we also estimated the relative clutch mass (RCM: clutch mass/female postparturient mass; Fitch 1987). All snakes were released within three to six days of capture. Data analysis.—We compared the SVL and ventral FIGURE 1. Earthsnake (Conopsis biserialis) from Ocoyoacac, México. (Photographed by Javier Manjarrez) and caudal scales between the sexes with a one-way ANOVA. Because head dimensions vary with body length, we used the SVL as a covariate in an analysis of fragmentation of the forest. We describe some aspects covariance (ANCOVA) to compare head dimensions of the ecology of C. biserialis from the State of Mexico. between the sexes. We used the Spearman test to explore Specifically, we describe the temporal abundance, body the relationship between the frequency of snakes size, reproductive condition, body temperature, and diet. captured and the rainfall and temperature, and between We compare our data with other fossorial-snake species. the body and under-rock temperature. We tested for sexual differences in the frequency of the snakes captured MATERIALS AND METHODS and the contained, identifiable prey with the Chi-square test (χ2). Statistical tests were made using STATISTICA Study site.—The population of C. biserialis we studied 8.0 (StatSoft Inc., Tulsa, Oklahoma, USA). We present was located at Ocoyoacac, 18 km S of Toluca City descriptive statistics as the mean ± 1 SD, we set alpha at (19°12’N, 99°19´W) at an elevation of 2,800 m. This 0.05, and we tested all data for normality. site is within the Mexican Transvolcanic Axis, characterized by a temperate forest surrounded by a RESULTS mosaic of agricultural land and small reforested patches. The specific habitat of the study site consisted of pine- Temporal abundance.—Conopsis biserialis had a oak forest (Pinus leiophylla-Quercus crassipes) with unimodal abundance pattern (Fig. 2). During both years, grasslands (Vázquez 2001). Rainfall averages 242 we captured 84 snakes; 41 in 2003 (49%) and 43 in 2004 mm/y, most falling June to September. The mean annual (51%). The annual period of activity was eight to nine environmental temperature is 13.1° C (García 1988). months. The first date we captured a snake was 7 February in 2003 and 12 February in 2004. The last Data collection.—We visited the approx. 1 km2 study snakes we captured were on 11 October in 2003 and 5 site biweekly from February 2003 to December 2004 September in 2004. We captured 87% of the snakes in from 0900 to 1500, although visits were less frequent the rainy season (June to September). The frequency of during winter (November to January) when snakes are captured snakes was highly correlated with monthly usually inactive. We looked for snakes under rocks or rainfall (rS = 0.85, df = 10, P < 0.001; n = 12; Fig. 2), fallen branches, on grassland, and in ground burrows. but not with the monthly temperature (rS = 0.37, df = 10, We collected snakes by hand, and we recorded the body P = 0.23; n = 12). temperature and under-rock temperature using a Schultheis thermometer (Miller & Weber Inc., Body size.—Most snakes we found during fieldwork Ridgewood, New York, USA). Except for pregnant (63%; n = 53) were adults (> 190 mm SVL,), averaging females, we obtained stomach contents by making the 235 ± 31 mm SVL (190–330 mm). Juveniles averaged snakes regurgitate. As an alternative method of diet 132 ± 29 mm SVL (96–185 mm, n = 31). Adult females determination, we obtained feces by manually pressing were slightly, but not significantly, larger than adult the anterior-vent region (Fitch 1987). The feces were males (F1,51 = 0.93, P = 0.33; Fig. 3). Adult males had washed with water to eliminate plant debris and soil, tails that measured 15% longer than females (F1,51 = clarified with 10% potassium hydroxide, and preserved 7.61, P = 0.008). Males also had significantly longer in 70% alcohol. heads than females (F1,82 = 11.81, P = 0.001; Fig. 3), but 365 Castaneda-Gonzalez et al.—Ecology of a Population of the Earthsnake Conopsis biserialis 300 30 Jerusalem Crickets (Orthoptera: Stenopelmatidae), ) 250 25 Lepidoptera larvae, and hyaline eggs. Individual mm ( SNAKES samples of 43 feces contained larvae of four items 200 20 OF identified as Lepidoptera, Arachnids, Hymenoptera, and 150 15 Orthoptera (Table 1). Adult insect items were present in RAINFALL 100 10 10% of feces and were from Lepidoptera, Scorpionida, NUMBER Acaridae, and Diptera (Table 1). In addition, the feces 50 5 also contained earth and plant debris, including moss, 0 0 roots, and grass.